In the object detection device, the plurality of object detection units output a score indicating a probability that a predetermined object exists, for each partial region set with respect to image data inputted. The weight computation unit computes a weight for each of the plurality of object detection units by using weight computation parameters and based on the image data. The weights are used when the scores outputted by the plurality of object detection units are merged. The weight redistribution unit changes the weight for a predetermined object detection unit, among the weights computed by the weight computation unit, to 0 and output the weights. The merging unit merges the scores outputted by the plurality of object detection units for each of the partial regions, by using the weights computed by the weight computation unit and including the weight changed by the weight redistribution unit. The loss computation unit computes a difference between a ground truth label of the image data and the merged score merged by the merging unit as a loss. Then, the parameter correction unit corrects the weight computation parameters so as to reduce the loss.

A method generates an image processing model to calculate a virtually stained image from a microscope image. The image processing model is trained using training data comprising microscope images as input data into the image processing model and target images that are formed via chemically stained images registered locally in relation to the microscope images. The image processing model is trained to calculate virtually stained images from the input microscope images by optimizing an objective function that captures a difference between the virtually stained images and the target images. After a number of training steps, at least one weighting mask is defined using one of the chemically stained images and an associated virtually stained image calculated after the number of training steps. In the weighting mask, one or more image regions are weighted based on differences between locally corresponding image regions in the virtually stained image and in the chemically stained image. Subsequent training considers the weighting mask in the objective function.

A circuitry of an image processing device divides a first image into a plurality of regions, extracts a feature point from each of the regions, tracks the feature point among a plurality of images to detect a motion vector, estimates a notable target of the first image, calculates the priority level of setting of a tracking feature point for each of the regions for tracking motion of the notable target, and sets the tracking feature point to any of the regions based on the priority level.

Methods, systems, and computer program products for static image segmentation are provided herein. A method includes segmenting an image containing a target object into multiple regions; analyzing video content containing the target object to determine a similarity metric across the multiple segmented regions based on information associated with the multiple segmented regions; and applying the similarity metric to the image to identify two or more of the multiple segmented regions as being portions of the target object.

A method, computer system, and a computer program product for identifying sections in a document based on a plurality of visual features is provided. The present invention may include receiving a plurality of documents. The present invention may also include extracting a plurality of content blocks. The present invention may further include determining the plurality of visual features. The present invention may then include grouping the extracted plurality of content blocks into a plurality of categories. The present invention may also include generating a plurality of closeness scores for the plurality of categories by utilizing a Visual Similarity Measure. The present invention may further include generating a plurality of Association Matrices on the plurality of categories for each of the received plurality of documents based on the Visual Similarity Measure. The present invention may further include merging the plurality of categories into a plurality of clusters.

The present invention is directed to implementing at least one of speed-up of detection processing and reduction of misdetection. An information processing apparatus includes an acquisition unit configured to acquire a captured image, a first setting unit configured to set a plurality of detection areas of an object for the captured image, a second setting unit configured to set a condition for detecting an object on a first detection area and a second detection area set by the first setting unit, wherein the condition includes a detection size in the captured image, and a detection unit configured to detect an object satisfying the detection size set by the second setting unit from the plurality of detection areas set by the first setting unit.

Systems, methods, and non-transitory media are provided for providing spatiotemporal recycling networks (e.g., for video segmentation). For example, a method can include obtaining video data including a current frame and one or more reference frames. The method can include determining, based on a comparison of the current frame and the one or more reference frames, a difference between the current frame and the one or more reference frames. Based on the difference being below a threshold, the method can include performing semantic segmentation of the current frame using a first neural network. The semantic segmentation can be performed based on higher-spatial resolution features extracted from the current frame by the first neural network and lower-resolution features extracted from the one or more reference frames by a second neural network. The first neural network has a smaller structure and/or a lower processing cost than the second neural network.

A method for correlating at least part of a candidate image (Ican) with at least one reference image, includes the following steps: a) implementing a relational repository (R) comprising at least: an ordered list of relational descriptors, at least one computing mode to be applied to the images in order to determine descriptors of these images, and a mode for determining the degree of similarity between two descriptors, b) implementing, for each reference image, a reference list that comprises the positions, referred to as reference points of interest, in the reference image, of descriptors of the reference image that are similar to relational descriptors from a relational repository compatible with the relational repository implemented in step a), which reference list is ordered on the basis of the order of this compatible relational repository, c) determining, in the candidate image, descriptors of the candidate image that are computed in line with each descriptor computing mode of the relational repository implemented in step a), and determining the position of each of these descriptors in the candidate image, d) determining the degree of similarity, determined in line with the determination mode of the relational repository implemented in step a), between each descriptor of the candidate image and each relational descriptor of the relational repository implemented in step a), e) determining a candidate list that comprises the positions, referred to as candidate points of interest, in the candidate image, of the descriptors of the candidate image exhibiting the greatest similarity with the relational descriptors of the relational repository implemented in step a), which candidate list is ordered on the basis of the order of this relational repository, f) processing the candidate list with respect to each reference list on the basis of the order of the candidate and reference lists.

An information processing apparatus includes a processor configured to: acquire information indicating a size of an external shape of each of characters in data of a first image and data of a second image that are used for comparison; and determine a presence or absence of a fault in the data of the second image with respect to each of the characters with reference to a degree of a difference that is between the data of the second image and the data of the second image and is detected in accordance with a detection condition varying in response to the size of the external shape of each of the characters.

A wearable device for use in immersive reality applications is provided. The wearable device includes eyepieces to provide a forward-image to a user, a first forward-looking camera mounted on the frame and having a field of view, a processor configured to identify a region of interest within the forward-image, and an interface device to indicate to the user that a field of view of the first forward-looking camera is misaligned with the region of interest. Methods of use of the device, a memory storing instructions and a processor to execute the instructions to cause the device to perform the methods of use, are also provided.